From: Will Deacon <will.deacon@arm.com>
To: "André Hentschel" <nerv@dawncrow.de>
Cc: "linux-arch@vger.kernel.org" <linux-arch@vger.kernel.org>,
Russell King - ARM Linux <linux@arm.linux.org.uk>,
"linux-arm-kernel@lists.infradead.org"
<linux-arm-kernel@lists.infradead.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"gregkh@linuxfoundation.org" <gregkh@linuxfoundation.org>,
Jonathan Austin <Jonathan.Austin@arm.com>
Subject: Re: [PATCHv3] arm: Preserve the user r/w register TPIDRURW on context switch and fork
Date: Wed, 8 May 2013 09:57:53 +0100 [thread overview]
Message-ID: <20130508085753.GA15568@mudshark.cambridge.arm.com> (raw)
In-Reply-To: <51896934.5080803@dawncrow.de>
Hi Andre,
On Tue, May 07, 2013 at 09:51:00PM +0100, André Hentschel wrote:
> From: =?UTF-8?q?Andr=C3=A9=20Hentschel?= <nerv@dawncrow.de>
Might just be my mailer, but you should check that your name is intact here
otherwise the git log will be mangled.
> Since commit 6a1c53124aa1 the user writeable TLS register was zeroed to
> prevent it from being used as a covert channel between two tasks.
>
> There are more and more applications coming to WinRT, Wine could support them,
> but mostly they expect to have the thread environment block (TEB) in TPIDRURW.
>
> This patch preserves that register per thread instead of clearing it.
> Unlike the TPIDRURO, which is already switched, the TPIDRURW
> can be updated from userspace so needs careful treatment in the case that we
> modify TPIDRURW and call fork(). To avoid this we must always read
> TPIDRURW in copy_thread.
>
> Signed-off-by: André Hentschel <nerv@dawncrow.de>
> Signed-off-by: Will Deacon <will.deacon@arm.com>
> Signed-off-by: Jonathan Austin <jonathan.austin@arm.com>
[...]
> diff --git a/arch/arm/include/asm/tls.h b/arch/arm/include/asm/tls.h
> index 73409e6..22756ab 100644
> --- a/arch/arm/include/asm/tls.h
> +++ b/arch/arm/include/asm/tls.h
> @@ -2,27 +2,30 @@
> #define __ASMARM_TLS_H
>
> #ifdef __ASSEMBLY__
> - .macro set_tls_none, tp, tmp1, tmp2
> +#include <asm/asm-offsets.h>
> + .macro switch_tls_none, base, tp, tpuser, tmp1, tmp2
> .endm
>
> - .macro set_tls_v6k, tp, tmp1, tmp2
> + .macro switch_tls_v6k, base, tp, tpuser, tmp1, tmp2
> + mrc p15, 0, \tmp2, c13, c0, 2 @ get the user r/w register
> mcr p15, 0, \tp, c13, c0, 3 @ set TLS register
> - mov \tmp1, #0
> - mcr p15, 0, \tmp1, c13, c0, 2 @ clear user r/w TLS register
> + mcr p15, 0, \tpuser, c13, c0, 2 @ and the user r/w register
> + strne \tmp2, [\base, #TI_TP_VALUE + 4] @ save it
Why is this conditional?
> .endm
>
> - .macro set_tls_v6, tp, tmp1, tmp2
> + .macro switch_tls_v6, base, tp, tpuser, tmp1, tmp2
> ldr \tmp1, =elf_hwcap
> ldr \tmp1, [\tmp1, #0]
> mov \tmp2, #0xffff0fff
> tst \tmp1, #HWCAP_TLS @ hardware TLS available?
> - mcrne p15, 0, \tp, c13, c0, 3 @ yes, set TLS register
> - movne \tmp1, #0
> - mcrne p15, 0, \tmp1, c13, c0, 2 @ clear user r/w TLS register
> streq \tp, [\tmp2, #-15] @ set TLS value at 0xffff0ff0
> + mrcne p15, 0, \tmp2, c13, c0, 2 @ get the user r/w register
> + mcrne p15, 0, \tp, c13, c0, 3 @ yes, set TLS register
> + mcrne p15, 0, \tpuser, c13, c0, 2 @ set user r/w register
> + strne \tmp2, [\base, #TI_TP_VALUE + 4] @ save it
> .endm
>
> - .macro set_tls_software, tp, tmp1, tmp2
> + .macro switch_tls_software, base, tp, tpuser, tmp1, tmp2
> mov \tmp1, #0xffff0fff
> str \tp, [\tmp1, #-15] @ set TLS value at 0xffff0ff0
> .endm
> @@ -31,19 +34,31 @@
> #ifdef CONFIG_TLS_REG_EMUL
> #define tls_emu 1
> #define has_tls_reg 1
> -#define set_tls set_tls_none
> +#define switch_tls switch_tls_none
> #elif defined(CONFIG_CPU_V6)
> #define tls_emu 0
> #define has_tls_reg (elf_hwcap & HWCAP_TLS)
> -#define set_tls set_tls_v6
> +#define switch_tls switch_tls_v6
> #elif defined(CONFIG_CPU_32v6K)
> #define tls_emu 0
> #define has_tls_reg 1
> -#define set_tls set_tls_v6k
> +#define switch_tls switch_tls_v6k
> #else
> #define tls_emu 0
> #define has_tls_reg 0
> -#define set_tls set_tls_software
> +#define switch_tls switch_tls_software
> #endif
>
> +#ifndef __ASSEMBLY__
> +static inline unsigned long get_tlsuser(void)
> +{
> + if (has_tls_reg && !tls_emu)
> + {
> + unsigned long t;
> + __asm__("mrc p15, 0, %0, c13, c0, 2" : "=r" (t));
> + return t;
> + }
> + return 0;
This isn't standard kernel coding style. Please do something like:
static inline unsigned long get_tlsuser(void)
{
unsigned long reg = 0;
if (has_tls_reg && !tls_emu)
asm("mrc p15, 0, %0, c13, c0, 2" : "=r" (reg));
return reg;
}
Will
next prev parent reply other threads:[~2013-05-08 8:57 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-05-07 20:51 [PATCHv3] arm: Preserve the user r/w register TPIDRURW on context switch and fork André Hentschel
2013-05-08 8:57 ` Will Deacon [this message]
2013-05-08 8:57 ` Will Deacon
2013-05-08 17:41 ` André Hentschel
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20130508085753.GA15568@mudshark.cambridge.arm.com \
--to=will.deacon@arm.com \
--cc=Jonathan.Austin@arm.com \
--cc=gregkh@linuxfoundation.org \
--cc=linux-arch@vger.kernel.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux@arm.linux.org.uk \
--cc=nerv@dawncrow.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox